Mitchell Creek Watershed Hydrologic Study Dave Fongers Hydrologic Studies Unit Land and Water Management Division Michigan Department of Environmental Quality September 19, 2007 Table of Contents Summary......................................................................................................................... 1 Watershed Description .................................................................................................... 2 Hydrologic Analysis......................................................................................................... 8 General ........................................................................................................................ 8 Mitchell Creek Results.................................................................................................. 9 Tributary 1 Results ..................................................................................................... 11 Tributary 2 Results ..................................................................................................... 15 Recommendations ..................................................................................................... 18 Stormwater Management .............................................................................................. 19 Water Quality ............................................................................................................. 20 Stream Channel Protection ........................................................................................ 21 Flood Protection ......................................................................................................... 25 References.................................................................................................................... 25 Appendix A: Mitchell Creek Hydrologic Analysis Data.................................................A-1 Appendix B: Mitchell Creek Hydrologic Parameters....................................................A-3 Appendix C: Glossary...................................................................................................A-6 This Nonpoint Source (NPS) Pollution Control project has been funded wholly by the United States Environmental Protection Agency through a Part 319 grant to the Michigan Department of Environmental Quality. This study is in support of a NPS grant, 2005-9119, to the Muskegon River Watershed Assembly. The contents of the document do not necessarily reflect the views and policies of the EPA, nor does the mention of trade names or commercial products constitute endorsement or recommendation for use. For more information, go to www.michigan.gov/deqnps. The cover is a 2005 aerial photo of the Mitchell Creek Watershed. For comments or questions relating to this document, contact Dave Fongers at: MDEQ, LWMD, P.O. Box 30458, Lansing, MI 48909 [email protected] 517-373-0210 Summary This hydrologic study of the Mitchell Creek watershed was conducted by the Hydrologic Studies Unit (HSU) of the Michigan Department of Environmental Quality (MDEQ) to better understand the watershed's hydrologic characteristics and reported continued channel instability subsequent to a Nonpoint Source (NPS) streambank stabilization project, 1999-0037. The watershed study has three scenarios corresponding to land cover in 1978 and 2006. Scenarios A and B simulate the actual condition of the watershed. Scenario C is hypothetical and is intended only for comparison. A. 1978 conditions B. 2006 conditions with a 0.05 cfs/acre release rate for new development C. 2006 conditions without a 0.05 cfs/acre release rate General land use trends for the watershed are illustrated in Figure 1. Additional land use information is provided in the Watershed Description section and in Appendix A of this report. The hydrologic modeling quantifies changes in stormwater runoff from 1978 to 2006 due to these land use changes. The dominant trend is urbanization and increased imperviousness near Big Rapids and Interstate 131. The associated increased runoff is managed by local stormwater ordinance. For more recent developments, runoff from a 50 percent chance (2-year), 24-hour storm event is limited to a maximum release rate of 0.05 cubic feet per second per acre (cfs/acre). Relatively modest, but frequent, storm events, such as the 50 percent chance storm, have more effect on channel form than extreme flood flows. Unless properly managed, increases in runoff from 1- to 2-year storms increase channel-forming flows, which increase streambank and bed erosion as the stream enlarges to accommodate the higher flows. Detailed discussion of the results is in the Hydrologic Analysis section of this report. The modeling indicates that the 0.05 cfs per acre standard in the stormwater management ordinance is helping protect Mitchell Creek and its tributaries from detrimental flow impacts of land use changes. Channel-forming flows near the mouth of Mitchell Creek have been not significantly changed by the addition of rate controlled developments within the watershed. However, refinements to the stormwater ordinance may help better protect smaller tributaries. These refinements could include 24-hour extended detention of runoff from 1-year storms or provision for retention and infiltration of additional stormwater runoff through Low Impact Development (LID) practices. Mitchell Creek Watershed Hydrologic Study 12/18/2007 page 1 Figure 1: Land Use Comparison, Mitchell Creek Watershed Watershed Description The 12.9 square mile Mitchell Creek watershed, Figure 2, outlets to the Muskegon River at Big Rapids and is located in Mecosta and Newaygo Counties. The watershed drainage area to the streambank stabilization project site, located just downstream of Northland Drive, is 12.7 square miles. One 0.6 square mile area of the watershed is defined as non-contributing, meaning it does not contribute surface runoff during flood events. This study divides the watershed into six subbasins, as shown in Figure 3. The watershed delineation changes in two places, Figure 4, from 1978 to 2006 because of earthwork associated with land use changes. Surface runoff volumes and flows were modeled using HEC-HMS 3.1.0 and the runoff curve number technique. This technique, developed by the Natural Resources Conservation Service (NRCS) in 1954, represents the runoff characteristics from the combination of land use and soil data as a runoff curve number. The technique, as adapted for Michigan, is described in “Computing Flood Discharges For Small Ungaged Watersheds (Sorrell, 2003). The runoff curve numbers were calculated using Geographic Information Systems (GIS) technology from the digital land use and soil data shown in Figures 5 through 8. The land use map depicting MDEQ GIS data for 1978 is shown in Figure 5. The 2006 land use map, Figure 6, is based on HSU’s analysis of 2005 and 2006 aerial photos. Mitchell Creek Watershed Hydrologic Study 12/18/2007 page 2 Housing density is a part of the curve number calculations. Based on the aerial photos, average residential lot size for both land use scenarios was specified as 1/3 acre in the vicinity of Big Rapids and 1/2 acre for the rest of the watershed. The NRCS soils data for the watershed is shown in Figures 7 and 8. Soil hydrogroups range from A to D, with A indicating well-drained, high infiltration soils and D indicating poorly-drained, high runoff soils. Where the soil is given a dual classification, B/D for example, the soil hydrogroup was selected based on land use. In these cases, the soil hydrogroup is specified as D for natural land uses, or the alternate hydrogroup (A, B, or C) for developed land uses. The differences in resolved soil hydrogroups from 1978 to 2006 are minor. The runoff curve numbers, calculated from the soil and land use data, are listed in Appendix B. The areas that developed since the 1978 scenario and were affected by the 0.05 cfs/acre standard were modeled as separate elements. An impervious area for each of these developed areas was assigned based on the land use GIS data, Figures 5 and 6, and Table 1. The imperviousness values for residential, commercial, and industrial land uses are from the NRCS (NRCS, 1986). The pervious portion of the drainage area was assigned a curve number of 30. Table 1: Imperviousness Table for Impervious Area Analysis GIS Class Description Imperviousness (percent) 1 Residential 38* 2 Commercial 85 3 Industrial 72 4 Road, Utilities 95 5 Gravel Pits 0 6 Outdoor Recreation 0 7 Cropland 0 8 Orchard 0 9 Pasture 0 10 Openland 0 11 Forests 0 12 Open Water 0 13 Wetland 0 * assumed population density of 250 to 1,000 people per square mile The time of concentration, which is the time it takes for water to travel from the hydraulically most distant point in the watershed to the design point, was calculated from the USGS quadrangles. The same time of concentration values was used in both land use scenarios. Storage coefficients were set equal to the times of concentration because there is little ponding within the watershed. Parameters are detailed in Appendix B. The design rainfall value used in this study is 2.27 inches, corresponding to the 50 percent chance (2-year) 24-hour storm, as tabulated in Rainfall Frequency Atlas of the Midwest, Bulletin 71, Midwestern Climate Center, 1992, pp. 126-129. This storm was selected because runoff from the 50 percent chance storm can be associated with channel-forming flows. Mitchell Creek Watershed Hydrologic Study 12/18/2007 page 3 Muskegon River Watershed Mitchell